Leonardo da Vinci may have lived 500 years ago, but he was well ahead of his time. Best known for painting the Mona Lisa, he also sketched a design for the first autonomous car. This is just one of the many fascinating tales that the history of self-driving cars is peppered with. Not forgetting the “phantom cars” of New York…
The Renaissance was an era of radical change. Probably the best known figure of the period is Leonardo da Vinci – painter, scientist, inventor and philosopher, remembered first and foremost for the Mona Lisa, history’s most famous painting. Many of his inventions were extremely futuristic. For instance, he designed the world’s first parachute, a diving suit, and various flying machines, hundreds of years before the first human succeeded in taking to the skies. But few people are aware that this polymath also pioneered a major current trend in the automotive industry – when he designed the world’s first self-driving car.
In 1478, da Vinci sketched a four-wheeled carriage powered only by a coiled spring. Considered the world’s first depiction of a self-propelled vehicle, his drawing can safely be assumed to be the starting point in the history of autonomous driving. Over 500 years were to pass before anyone could tell whether da Vinci’s design was actually roadworthy, because every attempt to create a working model based on the sketch was doomed to fail. It was only in 2004 that the Galileo Museum for the History of Science in Florence was able to crack Da Vinci’s car code and turn his design into a fully functional prototype. Da Vinci was not only the first person to investigate the possibility of autonomous driving – for many years he was also the only one.
In fact. the next milestone was the “Linrrican Wonder” of 1925, a “miracle car” that drove through traffic jams in New York as if steered by a ghost. This model was one of the so-called "phantom cars” of New York that were popular in the 1920s and 1930s, built to counteract enduring skepticism regarding the safety of road traffic. These phantom cars, however, were not in fact self-driving, as a human operator – rather than a computer – steered them remotely by radio control.
In 1933, America came much closer to creating a truly autonomous vehicle – but for the air rather than on the ground. US firm Sperry Gyroscope Company developed “Mechanical Mike”, the first aircraft autopilot. It enabled Wiley Post to become the first pilot to fly solo around the world, a journey he completed in seven days, 18 hours and 49 minutes. That same decade, US carmaker General Motors presented a vision of the car of the future at the 1939 New York World’s Fair. Its Futurama exhibit featured radio-controlled vehicles powered by electromagnetic fields. This, however, was to remain a vision for the time being.
The postwar years brought forth a few more futuristic visions. They included the Chevrolet Firebird III from the 1950s, a highly eccentric GM concept car that resembled a cross between a jet plane and a sports car. As well as its gas turbine propulsion, anti-lock brakes and touchscreen, the car was fitted with an “auto-guide“ system – a forerunner to today’s autonomous driving technologies. Coils fitted to the front wheels received signals from cables in the road, keeping the futuristic car automatically on the right track. Two years later in the U.K., the Transport Research Laboratory tested the same technology and modified a Citroen DS to follow cables in the road using magnetic sensors. Interesting as this development was, it had a short lifespan. Cars today don’t need power coils and electromagnetic fields to find their way around.
At the same time, systems that genuinely advanced the field of autonomous driving were also being developed. For instance, an extremely bumpy car ride with his chauffeur prompted US engineer Ralph Teetor to develop the predecessor to today’s cruise control systems. In 1945, Teetor registered the patent for his invention, and in 1958, “cruise control” was first installed in a vehicle – a Chrysler Imperial. In 1962, Mercedes became the first European carmaker to follow suit by manufacturing vehicles with a cruise control system. And cruise control isn’t the only technology that began to take off in the post-war period. From the end of the 1950s, the USA and USSR were locked in a fierce race to the Moon, and the car industry benefited from some of the resulting technologies. With not terrestrial but lunar travel in mind, researchers at Stanford University worked on developing a vehicle for the moon landing. The “Stanford Car” was designed to be remotely controlled from Earth, but was also fitted with a camera to compensate for the delay between transmitter and receiver due to the vast distance. This camera made it possible for the vehicle to follow a white line fully autonomously. Its autonomous capabilities were also tested on Earth. The parallels to contemporary driver assistance systems such as Lane Assist are clear to see, as these too typically use cameras for orientation.
It wasn’t just the Americans, though, who were intrigued by the potential of autonomous driving; around 20 years after the moon landing, a German robotics specialist made a major contribution to the field. In the 1980s, Ernst Dieter Dieckmanns, a professor at the Bundeswehr University in Munich, initiated the “Prometheus” project dedicated to researching and developing self-driving vehicles. In 1994, the researchers managed to get a modified Mercedes van to drive autonomously for over 1,000 kilometers on the autobahn – at 130km/h! The Mercedes could switch lane by itself and even overtake other vehicles autonomously. Adaptive Cruise Control and the Autonomous Emergency Braking, now established technologies, were developed during this research project.
Autonomous driving made another leap forward with the founding of DARPA, an agency of the US Department of Defense specializing in self-driving cars. The agency launched the DARPA Grand Challenge to raise the public profile of autonomous vehicles and promote research and development in the field. Since 2010, tremendous advances in autonomous driving have been made. In 2012 in Las Vegas, a self-driving car was first approved for use on public roads. In 2015, automotive manufacturer Tesla introduced a software update for its Model S, which includes a semiautonomous autopilot. The Tesla can maintain a safe distance from the vehicle in front, stay within its lane and switch lanes, and detect vacant parking spaces and red traffic lights – all fully autonomously. And starting this year, robocabs have become a regular feature on the streets of Las Vegas. These, however, are not fully autonomous – they may brake, steer and accelerate by themselves, but there is still a driver at the wheel in case of emergencies and to comply with the law. After all, who is liable if an autonomous car causes an accident?
Once this legal question has been settled, there’s nothing to stop fully autonomous vehicles from taking to the roads. The Fraunhofer Institute forecasts that self-driving cars will become a regular feature in road traffic from the year 2030. How amused would Leonardo be to know that humanity has taken 500 years to crack his code –and we’re not quite there yet.